The ongoing development of information acquisition technique resulted in high-dimensional and large-scale data, it enormously challenges the data mining. Aiming at low efficiency and high time cost of K-nearest neighbor classification algorithms in high-dimensional data, an Improved k-nearest neighbor algorithm based on weight search tree (WSTKnn) for high-dimensional classification was proposed in this paper. The algorithm selects some attributes as nodes to construct a search tree according to the weight of feature attributes. The search tree divides the data set into different matrix regions. Unknown samples need to find the search tree to obtain the most "similar" matrix region, and only calculate the distance from the data contained in the matrix area. thus, reduce data size to reduce time complexity. And discussed the Minkowski Distance that would be most suitable for distance measurement of high-dimensional data. Simulation experiments on 6 standard high-dimensional data show that the classification time of WSTKnn has better performance than K-nearest neighbor, Decision Tree and SVM, the classification time is significantly reduced and the classification accuracy is better than other algorithms. WSTKnn has better performance on the classification of high-dimensional data, which is expected to give some references for solving the related problem of high-dimensional data.